70 T
effect of propofol on the extracellular glutamate level Sycopel International, Ltd., Boldon, Tyne and Wear, UK
during cerebral ischemia remains to be elucidated. In this for continuous measurement of extracellular glutamate
study, we tested the hypothesis that propofol delivered concentration as described previously in detail [3,26]. In
directly to the CNS produced a reduction in extracellular brief, the microdialysis biosensor was made up of a
glutamate level during global ischemia and the resultant platinum working electrode, a silver silver chloride coun-
neuroprotection in the hippocampal CA1 region of the rat. ter electrode, a silver reference electrode, and two vitreous
silica tubes for changing the solution inside the biosensor, all of which were set inside a hollow semi-permeable
2. Materials and methods membrane, 230 mm in outside diameter. The biosensor was
filled and perfused with phosphate buffered saline PBS 2.1. Reagents
pH 7.4 mM: NaCl, 146; KCl, 2.7; Na HPO E H O, 4.3;
2 4
7 2
KH PO , 1.4; CaCl , 2.4 by a perfusion pump EP60;
2 4
2
Glutamate oxidase, propofol, and Intralipid were ob-
Eicom, Kyoto, Japan. O-phenylenediamine was elec- tained from Yamasa Chiba, Japan, Aldrich Milwaukee,
tropolymerized onto the platinum electrode in PBS bub- WI, USA, and Pharmacia AB Stockholm, Sweden,
bled with 100 nitrogen and stirred to avoid interference respectively. All other reagents were purchased from
from electroactive molecules such as ascorbate. The cur- Nacalai Tesque Kyoto, Japan.
rent from hydrogen peroxide produced by glutamate oxidation was detected amperometrically on the platinum
2.2. Subjects, preparation, and ischemia electrode at 1650 mV by an electrochemical detector EPS
800, Eicom and recorded on a polygraph in real time. All animal care procedures in this study were performed
After stabilization of the current at 1650 mV, the biosensor according to the Guidelines for Animal Experiments of the
was perfused with glutamate oxidase 0.05 U ml dis- Kumamoto University School of Medicine. The animal
solved in PBS at a rate of 0.2 ml min. The calibration in care and use committee for our institute approved the study
vitro was performed by placing the tips in the PBS protocol. We used adult male Wistar rats weighing between
containing cumulated concentrations of
L
-glutamate. The 230 and 290 g. The animals were given free access to food
current that developed on the platinum electrode increased and water. At the time of the experiment, they were
linearly with
L
-glutamate concentrations up to at least 300 anesthetized with 4 halothane in nitrous oxide oxygen
mM. The sensitivity and response time of the microdialysis F O ; 0.33, after which their tracheas were intubated and
biosensors used were around 0.15 nA mM and within 20 s
I 2
they were connected to a rodent ventilator 7025; Ugo for a 90 steady state response, respectively. The micro-
Basile, Camerio, Italy. They were mechanically ventilated dialysis biosensors were also calibrated for
L
-glutamate in tidal volume, 10 ml kg; frequency, 60 min with 1
different concentrations 10 and 30 of Intralipid, which halothane in nitrous oxide oxygen. The pericranial tem-
was a vehicle for propofol in this study see Experimental perature was continuously monitored with a tissue implant-
design, to see whether Intralipid influenced on the cali- able thermocouple microprobe IT-14; Physitemp Instru-
bration curve. The biosensor was implanted stereotaxically ments
Inc., Clifton,
NJ, USA
and maintained
at on the left side of the hippocampal CA1 area 75 min
37.060.28C during surgery, brain ischemia, and recovery before the initiation of ischemia, and perfused with PBS
from anesthesia with the use of a temperature controller containing glutamate oxidase at a rate of 0.2 ml min to
TCAT-1A; Physitemp Instruments Inc. and a radiant initiate in vivo real-time measurement of glutamate. The
heating lamp. Electroencephalogram EEG from bilateral area under the curve defined as integrated increments of
needle electrodes was continuously monitored and re- glutamate concentration from the pre-ischemic baseline
corded. The tail artery was cannulated for monitoring of value between the beginning and 20 min after the end of
arterial pressure and analysis of arterial blood gases, ischemia was calculated from the actual trace using NIH
glucose, and hematocrit. Global forebrain ischemia was Image version 1.61 software.
achieved by the four-vessel occlusion method originally described by Pulsinelli and Brierley [27]. In brief, vertebral
2.4. Experimental design arteries were coagulated at the alar foramina of the first
cervical vertebra, and then each common carotid artery Rats were randomly assigned to one of four groups to be
was snared and tightened with a 3-0 silk suture by treated with: a propofol 3 mg kg n59; b propofol 10
applying a weight of 15 g. Forebrain ischemia was mg kg n510; c Intralipid n510; or d artificial
confirmed by isoelectric EEG. Reperfusion was initiated cerebrospinal fluid aCSF; mM: NaCl, 126.5; KCl, 2.4;
by cutting the carotid sutures. KH PO , 0.5; CaCl , 1.1; MgCl , 1.1; Na SO , 0.5 n5
2 4
2 2
2 4
9. Propofol was dissolved in Intralipid, a fat emulsion 2.3. Microdialysis biosensor
composed of 10 soya bean oil, 1.2 egg yolk lecithin and 2.5 glycerol, since it is poorly water-soluble. These
We used a microdialysis biosensor General 20-10-2-2; agents, 20 ml in volume and warmed at 378C, were
T . Yano et al. Brain Research 883 2000 69 –76
71
stereotaxically infused for 1 min into the bilateral cerebral ventricles 10 ml into each ventricle through burr holes 15
min prior to the initiation of ischemia. The glutamate analysis and anesthesia were terminated 30 min after the
initiation of reperfusion. Animals were directly monitored until they fully woke from the anesthesia, then their
tracheas were extubated.
2.5. Histology The rats surviving for 7 days were sacrificed under
pentobarbital anesthesia, then perfused transcardially with heparinized normal saline followed by 10 neutral buf-
fered formalin. The brains were removed and fixed in 10 neutral buffered formalin, processed and embedded in
paraffin. Five-micrometer-thick coronal sections were taken from the dorsal hippocampus and stained with cresyl
violet acetate. The number of histologically intact pyrami- dal cells with a distinct nucleus and nucleolus in a 1-mm
length of the middle portion of the CA1 subfield was counted on the right hippocampus by an inspector blinded
Fig. 1. A calibration curve of the microdialysis biosensor for glutamate in
to intervention using a microscope at 3400 magnification.
different concentrations of Intralipid. A linear regression line is shown
The location of the microdialysis electrode was ascertained
only for the plots from the calibration in Intralipid-free PBS.
in each 5 mm-thick coronal section stained with hemato- xylin and eosin. If the track of the electrode was outside
the CA1 region, the glutamate data was excluded. chemia was higher than those were at the baseline and
after the intracerebroventricular administration in all treat- 2.6. Statistics
ment groups P,0.05. Neither aCSF nor Intralipid administration into the
Values were expressed as the mean6S.D. and data were cerebral ventricles influenced the EEG before ischemia. A
analyzed by a one-way analysis of variance or repeated- pattern of burst suppression with an interval reaching
measures analysis of variance. The Student–Newman– approximately 10 s appeared within a few minutes after the
Keuls multiple comparison procedure was then used to intracerebroventricular injection in all rats given 3 or 10
determine which pairs of means differed. A chi-square test mg kg of propofol. Thereafter, the EEG was gradually
was used to identify differences in the survival rates. restored to the same pattern as that seen before propofol
Statistical significance was defined as a P value ,0.05. administration, and the pattern of burst suppression was no
longer observed in any of the propofol-treated rats by 15 min after the injection just prior to initiation of ischemia
3. Results Fig. 2.
